International collaborative efforts employing high-throughput sequencing methods have led to the characterization of genomic alterations arising in multiple cancers and tumor cell lines. Analyses of these data have highlighted the complexity of this mutational landscape and, in particular, they have revealed a strong cancer lineage dependence with respect to both, the mutational patterns and their phenotypic outcomes: the mechanisms driving tumors are dependent of the molecular particularities (gene expression program, signaling networks etc.) of the cell they are derived from.
Cells, tissues and organs obtain their specificities through the set of genes they express. This transcriptional program is executed by specific transcription factors that directly or indirectly bind DNA and control gene expression levels. Transcription factors are highly relevant in cancerogenesis. Indeed, transcription factors and transcription regulators are often oncogenic drivers, such as MYC, for instance.
We will deconvolute the complexity of the tumor lineage expression program by modeling transcription factor activity in order to identify genomic alterations relevant for the tumorigenesis of specific cancer lineages. This approach will be developed by exploiting existing datasets and data generated within the MILES consortium, in which numerous tumors have been characterized at the genomic and transcriptomic level.
This computational biology sub-project will integrate data of multiple cancer types. We will try to discriminate tumors based on their molecular status, which may allow improved diagnostics. We will integrate cellular specificities to contextualize the genomic alterations. Moreover, analyzing the impact of genetic alterations on transcription factor activity as a function of tumor type will provide important mechanistic insights for the improved classification and stratification of tumors. The project is structured around four principal aims: